2011/8/16 Thomas <th.vanderv...@gmail.com>:
> Hi everyone,
>
> I have been struggling with this, hopefully, simple problem now for
> quite sometime, What I want to do is:
>
> *) read a file line by line
> *) modify each line
> *) write it back to a different file
>
> This is a bit of sample code that reproduces the problem:
>
> ==========================================================
> (def old-data (line-seq (reader "input.txt")))
>
> (defn change-line
> [i]
> (str i " added stuff"))
>
> (spit "output.txt" (map change-line old-data))
> ==========================================================
> #cat "output.txt"
> clojure.lang.LazySeq@58d844f8
>
> Because I get the lazy sequence I think I have to force the execution?
> but where
> exactly? And how?
>
> Thanks in advance!!!
spit operates on strings, so therefore you have to turn your data into
one big string first. (spit implicitly calls str on its argument, but
this is not very useful.)
What you have is a sequence of string, so depending of what you want
to appear in the file you have multiple options:
For an arbitrary clojure data structure you can use pr-str to convert
it into the same format you get at the repl: (spit "output.txt"
(pr-str (map change-line old-data))). This can be useful to dump some
data to a file and will yield something like this:
("line1 added stuff" "line2 added stuff")
To simply write a file where each line corresponds to a string element
in the sequence, you can either build a new string with consisting of
the strings of the seq, each with a newline character appended to the
end, concatenated together and spit that, or you can use something
else that doesn't require you to build this monolithic string. Since
you used line-seq rather than slurp to read in the file, I will
instead demonstrate an other approach than spit:
(require '[clojure.java.io :as io])
(with-open [in (io/reader input-filename)
out (io/writer output-filename)]
(binding [*out* out]
(->> in (line-seq) (map change-line) (map println) (dorun))))
This consumes the sequence line by line and writes the lines to the
file. This solution only needs to have one line in memory at a time.
The spit approach would require one big string to be constructed, and
might not be very suited for big files. The code would output a file
like this:
line1 added stuff
line2 added stuff
So in general, use slurp together with spit or read-line (or its
line-seq variant) together with println.
// raek
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
